<div>Many thanks to Manuel and Teodoro.</div>
<div> </div>
<div>My last question (hopefully)...</div>
<div> </div>
<div>It is always best to keep the consistency between the E_xc used in the DFT calculation and that used in generating the pseudopotential. There really isn't a big difference for something like PW91 vs. PBE but HSE vs. PBE would cause problems. (I think Ge is one example of this, there are problably others.)
</div>
<div> </div>
<div>Does the GTH PP generator support HF exchange?<br> </div>
<div>Alternative, is it possible to somehow "convert" a pseudopotential from either fhi98pp or opium into the CP2K format. (I do understand that there are some very big differences between GTH and say HSC type pseudopotential).
<br> </div>
<div><span class="gmail_quote">On 8/1/07, <b class="gmail_sendername">Teodoro Laino</b> <<a href="mailto:teodor...@gmail.com">teodor...@gmail.com</a>> wrote:</span>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div style="WORD-WRAP: break-word">libint 1.1.2 works for me
<div><br> </div>
<div>teo</div>
<div>p.s.: in case of troubles with the compilation of the libint library you may want to point the problem to the libint team..</div>
<div><br>
<div>
<div><span class="e" id="q_11420f78c2a76e43_1">
<div>On 31 Jul 2007, at 23:18, Nichols A. Romero wrote:</div><br></span></div>
<blockquote type="cite">
<div><span class="e" id="q_11420f78c2a76e43_3">
<div>Manuel,</div>
<div></div>
<div>Which version of the libint library are you using? Had some trouble compiling 1.1.3<br> </div>
<div>Thanks,<br> </div></span></div>
<div>
<div><span class="e" id="q_11420f78c2a76e43_5"><span class="gmail_quote">On 7/30/07, <b class="gmail_sendername">mguidon</b> <<a onclick="return top.js.OpenExtLink(window,event,this)" href="mailto:manuel...@gmail.com" target="_blank">
manuel...@gmail.com</a>> wrote:</span> </span></div>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">
<div><span class="e" id="q_11420f78c2a76e43_7"><br>Hi.<br><br>The HSE functional is a hybrid functional. It consists of screened<br>exact Hartree-Fock exchange and two GGA-functionals (PBE and hole- <br>averaged screened PBE functional). The keywords are (see /tests/QS/
<br>regtest-hybrid/NE-hybrid-HSE06-lda.inp):<br><br>&XC_FUNCTIONAL<br>&XWPBE<br>SCALE_X -0.25<br>SCALE_X0 1.0<br>OMEGA 0.11<br>&END <br>&PBE<br>SCALE_X 0.0<br>SCALE_C 1.0<br>&END PBE<br>&END XC_FUNCTIONAL
<br>&HF<br>EPS_SCHWARZ 1.0E-10<br>MAX_MEMORY 10<br>FRACTION 0.25<br>SCREENING_TYPE SHORTRANGE<br>OMEGA 0.11<br>&END<br><br>As in the case of B3LYP the functional is quite expensive since exact<br>exchange has to be calculated. However, the keyword OMEGA decreases
<br>the cost for the Hartree-Fock part because it results in a better <br>screening. To get accurate results it is recommended to use OMEGA<br>values within the range of 0.1 and 0.2. (Higher value means better<br>screening means faster execution). In the current version, analytical
<br>derivatives up to second order are implemented which means that forces/ <br>stresses are available.<br>As with all calculations including Hartree-Fock exchange you need an<br>up-to-date g95 compiler together with the libint-library in order to
<br>run cp2k.<br><br>Cheers<br><br>Manuel Guidon<br><br><br><br></span></div><span class="q">Reston, VA<br>443-567-8328 (C)<br>410-278-2692 (O) <br><br><br></span></blockquote></div></blockquote></div><br> </div><span class="q">
<br><br></span></div><br></blockquote></div><br><br clear="all"><br>-- <br>Nichols A. Romero, Ph.D.<br>High Performance Technologies, Inc.<br>Reston, VA<br>443-567-8328 (C)<br>410-278-2692 (O)